new compute kernel design, less mem read delay

2026-04-07 01:04:05 +08:00
parent d702b3224c
commit 12a216d404
47 changed files with 225 additions and 145 deletions
@@ -4,6 +4,7 @@

 #include "sn_intrinsic.h"
 #include "sn_intrinsic.std.h"
+#include "sn_intrinsic.context.x86.h"
 #include <math.h>

 static const float quality_threshold = 1e-6f;